Smart Trends in Information Technology and Computer Communications

Machine learning methods majorly comprise of image processing and soft computing methods and are mainly responsible for automation. Wheat production is influenced by assorted varying factors. Sorting or grading of agricultural products influenced by computer varies product wise and even product variety wise which itself changes region wise. Grading for new varieties released by the agricultural scientists is the major concern as new varieties are produced by crossing existing varieties. For these varieties, proven optimized machine learning algorithms may give an adverse result. This paper introduces machine learning algorithm capable of classifying major 5 wheat cultivars cultivated in Gujarat region of India. Experimental data consist of 11 traits comprising of shape, color and morphological characteristics. After applying feature selection algorithm, 5 traits were considered and Levenberg-Marquardt back propagation was employed to classify above wheat cultivars which ensued to more than 90% overall accuracy.

A communication gap has always existed between Sign Languages and other Natural Languages. This paper aims to build a real-time autonomous system that can help bridge this communication gap. The present system captures the gestures using a webcam and recognizes the gesture being shown, mapping it to the corresponding English Letter, Numeric Digit and Special Characters. The authors have proposed American Sign Language with some minor modifications. The present form of ASL can be used to recognize all alphabets (A–Z), all numerals (0–9), Backspace, Blank Space. Some Special Characters have been included as well. The present system is built to recognize the finger-spelling component of the American Sign Language (ASL), and can be extended to recognize other sign languages as well.

Mobile Ad hoc network is a collection of mobile nodes. It has the property to change according to changing topology due to the lack of centralized control. The nodes act as a sender and receiver both and communicate with each other through various protocols. We have analyzed the energy factor of existing AODV protocol and have done improvement in it by modifying MAC layer of nodes. We have made the nodes to sleep after the predefined interval which leads to reduction in energy consumption and thus enhances the lifetime of mobile nodes.

Silicon-on-insulator (SOI) suffers with various drawbacks so Silicon-on-nothing (SON) has been the researchers recent target. The result shows that surface potential and electric field is maximum compare to Single Material Gate as well as Dual Material Gate SOI junctionless Transistor (DMG SOI JLT). Various targeted features comprise such as maximum on-state current, improved transconductance Gm, Gm/I_DS and reduced drain induced barrier lowering. In this paper is the work on the parametric effect of two layer gate stack (DGS) (High dielectric/Sio₂) on the Dual Material Gate (DMG) for SON Junctionless Transistor. The results obtained by the simulation for 40 nm channel length with work function as 4.77 and 4.1 eV with doping concentration (0.4 × 1018 cm−3).

In this paper two dimensional photonic crystal ring resonator based add/drop filter is proposed. The layout of the proposed filter includes bus waveguide and drop waveguide which is coupled with the quasi-square resonator. The filter has a square lattice with rods suspended in air type structure. The add/drop filter has 1500 nm add (off resonance) wavelength and 1567 nm drop (on resonance) wavelength with drop efficiency 95%. This type of filter is mainly used in CWDM system. The size of the filter is very compact 11.1*10 µm2. The designed filter has quality factor of 3134. The layout of filter is designed using a layout designer tool. OptiFDTD software is used for the designing and simulation of filter.

The paper provides wide range survey of techniques, methodologies and systems for Object localization in indoor environment space using BLE (Bluetooth Low Energy) technology. It also presents study to track moving smart objects and provides their comparison based on factors such as privacy, accuracy, and location type. One important problem in object localization using Bluetooth is to identify the position of BLE tags and produce accurate location results using certain algorithms. In this survey, theft prominent research directions are categorized, analyzed and discussed.

Huffman coding and Fourier Transform technique are the basic technique used in image compression. Fourier transform is very powerful technique compared to Huffman coding because Fourier transform has ability to use in multiorders. The purpose of this paper is to compressed digital images using Huffman coding and Fast Fourier Transform and compare the results of both techniques. In calculation of parameters Matlab tool required. These techniques are compared with respect to various parameters such as mean square error (MSE), Peak signal to noise ratio (PSNR), Compression ratio (CR) and Bits per pixel (BPP) for the various input image of different size and it involves the new method of splitting or dividing an input image into equal rows & columns and at final stage sum of all individual compressed images.

Image enhancement is the process of highlighting some characteristics and carrying out certain features in original image for any problem oriented applications. Two domains in which image enhancement can be done is either in spatial domain and frequency domain with good SNR and keeping original colors of image intact. Metaheuristic approach provides a very effective search and optimization approach which gives good results in comparison to traditional approaches. In this paper author improve number of pixels, so that more details of image can be visualized easily and accurately. Two Metaheuristic algorithms, cuckoo search and firefly algorithms are applied here to find out optimal solution which gives peak performance. Experimental results of both the algorithms are tabulated and compared, which shows that firefly algorithm gives better performance as compared to cuckoo search algorithm in terms of robustness, fitness function and convergence rate. Some hybridization of metaheuristic algorithms can also be applied to improve performance.

Biometric recognition is a way of recognizing people on the basis of their behavioral and physiological characteristics. Palm print recognition is a very popular biometric recognition method because of its stable line features, need of low cost capturing device, low resolution imaging and user friendliness. Palm print recognition has been area of interest for many researchers since last many years due to the unique and stable characteristics present in a Palm. Researchers have suggested various preprocessing, feature extraction and matching techniques for recognition of a Palm print. This paper discusses various stages of Palm print recognition and research work performed in field of Palm Print recognition.

This paper outlines the Plugin for various browsers for instant rejuvenation and translation of web pages into Indian languages. The websites with Hindi content are less than 0.1% of the total websites, similarly less than .01% for other Indian languages. While English content are more than 55% [13]. It is high time for realization to provide the information to the local uses into their local languages so, that they can take the advantage of various resource available on websites which would result in enhancement in communication and knowledge. The Plugin tool can be plugged in various browsers. On single clicks, the whole page gets translated and rebuild into the original format without losing any information and graphics, which makes it easy, convenient, and readable for the users. The methodologies used in our system are Extraction, Rebuilding and Translation Memory. Further, we will discuss the workflow among the processes and then concluded with experimental results that are obtained with this tool.

POS tagging, an ideal way to augment a corpus is an imperative abstraction for text mining. However with an increase in the amount of linguistic errors and distinctive fashion of language ambiguities, the data filtered by POS tagging is noisier. In this paper, probabilistic tagging and tagging based on Markov models are combined to estimate the association probabilities. Based on this combined approach, error estimation model is defined. Comparison study is made on different corpus available in NLTK such as Crubadan, Brown and INSPEC. The results obtained by the proposed methodologies show a drastic increase in the accuracy rate of about 98% when compared to the existing algorithms which shows an average of 96% accurate. The performance measure is plotted to calculate the error ratio across the maximum-likelihood estimation.

Recommendation systems have been successfully used by e-commerce and other similar sites for recommendation of relevant items to the user. However majority of these systems are based on web usage mining which does not consider the semantic knowledge underlying a website in the recommendation process and based solely on usage data. Hence researchers realized the importance of semantic knowledge and began to use it as part of usage data which is primarily used by personalization systems to enhance the quality of items being recommended. However several issues emerged during the process of integration. For effective personalization these issues need to be addressed. We discuss this aspect of integration process and also suggest some of the ways to resolve these issues and also discuss few methods of representing domain knowledge under different situations.

Due to complexity of software design, software services are becoming essential with privacy and security. Service-Oriented Designing (SOD) pattern is one of the reputable patterns used for developing secure, reliable and flexible software. Consequently, the use of SOD to develop the durable security of software is increasing. Security measurement has considerable importance in the context of SOD since it determines how the requirements for secure service should be achieved for duration. In this paper, the relationship between durability and secure software of service-oriented design is proposed. The paper begins with intending a set of durability concepts to measure the security and concludes with the relationship between service-oriented design properties and durability of secure software.

A fuzzy type image fusion method using various image compression techniques is described in this paper. This method shows fusion of fuzzy images and can be used for fusion of multi model image. It is concluded that fusion with advanced single levels offers better fusion quality. The future algorithm Curvelet transform is very straight forward, effortless to execute and can be extended its use in existent time applications. This method provides a comparative study between proposed & literature techniques and validation of the projected algorithm as Peak Signal to Noise Ratio (PSNR), Root Mean Square Error (RMSE).

Purpose: Like every industry, ICT applications have great impact on the Hospitality & Tourism industry. From social media to smart phones and automatic check-ins, Information and Communication Technology effects even the smallest areas of the industry. The utilization of Information and Communication Technology are evolving very rapidly and Hospitality and Tourism Industry have to acclimatize quickly to keep ahead in the competitive market.The uptake of recent technological advances has been slow in the Industry. Unless this can be improved in the next number of months and years, many Hospitality and Tourism related companies risk falling even further behind when the new wave of technological advances comes along. Businesses that fail to adapt to the changing technological environment, inevitably, end up failing and shutting down. Technology in the Hospitality and Tourism industry has advanced a lot in recent years and will continue to advance at an incredible pace for the foreseeable future. Recently the Tourism industry has been slow to respond to change but luckily many Tourism related companies have seen the error of their ways and are beginning to adapt to these new technologies such as apps and smartphones. Providing the industry can catch up before a new wave of technology comes along, it should then be in a position to get ahead of the game and be proactive rather than waiting for every other industry to pass it out. Although it is seen as a difficult industry in which to innovate, there are now and will always be ways to adapt new and emerging technologies to the Hospitality and Tourism sector.

Wireless networking is on the verge of boom these days. Many multinational companies are trading in spectrum to provide data and voice services over wireless and mobile space. This research paper is a comparison of various network optimization as well as compression technologies for wireless network and mobile network. Asynchronous Transfer Mode (ATM), Genetic Algorithms (GA) is compared. It is reviewed that how it can bring efficiency in the current wireless as well as mobile networking. This research paper also reviews loseless and lossy compression and its efficacy in wireless networking scenario. This research paper provides reader a fair and honest view of current compression and speed optimization technologies, so that reader could get an insight of wireless network scenario in the prevailing conditions.

Sensors are resource constrained and computing devices used in wireless networks. These networks are comprised of large numbers sensors deployed randomly over an area. The wireless sensor network (WSN) has a direct impact on human welfare as their application can be extended to military surveillance, environmental monitoring and to healthcare also. Security breaches might lead to grave consequences, so it is important to protect wireless sensor networks against such threats. The specific characteristics of wireless sensor networks make them vulnerable to attacks on their communication channels and their hardware. This research work discusses the open security issues in WSN. Section 2 focuses on the problem statement and later, existing security mechanisms are discussed. This research paper also present an discussion on the Cryptographic mechanisms, which can be employed to protect against some of the possible attacks such as eavesdropping and the injection of messages by the attacker is prevented by authentication.

The paper delves into a relative examination of various Image Segmentation Techniques and also evaluated the performance of Thresholding and Region Growing Image Segmentation methods based on Jaccard Similarity Coefficient (JSC), Dice Similarity Coefficient (DSC) and execution time.

Contrast enhancement and brightness preservation of the image are two important issues of image enhancement in research field now-a-days. The objective is to enhance the image uniformly over different parts of the image. General Histogram Equalization doesn’t control degree of enhancement of the image. To overcome this drawback, another variant of Histogram Equalization method namely Adaptive Bi-histogram Equalization using Threshold (ABHET) is being proposed. The proposed method undergoes three steps, such as: Histogram segmentation using threshold, Clipping of histogram using mean value of occupied intensity and histogram equalization of each sub-images. Finally all the sub-images are combined into one complete image. Simulation results show that ABHET outperforms other existing HE-based methods and different image quality measures such as: Peak signal to noise ratio (PSNR), Absolute Mean Brightness Error (AMBE) and Structural Similarity Index (SSIM) are being used to test the robustness of the proposed method in terms of enhancement of contrast and preservation of brightness.

Sentiment analysis becomes a very active research area in the text mining field. It aims to extract people’s opinions, sentiments, and subjectivity from the texts. Sentiment analysis can be performed at three levels: at document level, at sentence level and at aspect level. An important part of research effort focuses on document level sentiment classification, including works on opinion classification of reviews. This survey paper tackles a comprehensive overview of the last update of sentiment analysis at document level. The main target of this survey is to give nearly full image of sentiment analysis techniques at this level. In addition, some future research issues are also presented.

An enormous growth of the WWW has been instrumental in spreading social networks. Due to many-fold increase in internet users taking to online reviews and opinions, the communication, sharing and collaboration through social networks have gained importance. The rapid growth in web-based activities has led to generation of huge amount of unstructured data which accounts for over 80% of the information. Exploiting big data alternatives in storing, processing, archiving and analyzing this data becomes increasingly necessary.In this paper we propose a generalized approach to analyzing sentiments in big-data environment. The proposed model would serve to incorporate different supervised and un-supervised approaches to extraction, classification and scoring of opinions and sentiment words.

A machine learning based classification technique to diagnose lung CT scan images as cancerous or noncancerous is proposed in this paper. Lung cancer is regarded as one of the major fatal disease among the population throughout the world. Early diagnosis of lung cancer can be an important factor which can decrease the death rate among people. In large medical organizations manual inspection of CT scan, MRI images etc. puts a lot of workload on doctors and radiologist. An effective diagnosis technique can really reduce their efforts. CT (Computerized Tomography) scan images are used in medical field to analyze various parts of body. Grey scale CT scan images are used here as dataset, image preprocessing and feature of images are used. SVM classifiers are used for diagnosis. The main objective of this paper is to improve accuracy rate for lung cancer diagnosis by designing a hybrid SVM.

Multiple input multiple output (MIMO) control system conflicts with single input single output (SISO) control system in many ways because of various parameters. Two tank coupled interacting is one such system which has two inputs and two outputs and the control strategy used for it is not like SISO. Electrical control of the system mainly deals with control of supply to the pump which is the actual source of water to both the tanks. PID controller is generally used for controlling the liquid level. Basic PID controller has some demerits related with overshoot in the response and steady state error. This responses experimented on actual hardware are shown. Intelligent controller like fuzzy logic controller (FLC) when combined with PID can give better responses. These responses are validated and shown in this paper which is formulated in the form of simulation.

Several unconventional clustering problems have been defined recently with some preliminary solutions. This paper discusses one such approach useful for community detection in social networks. Multi Chromatic Correlation Clustering deals with identifying groups of data objects based on the category of relation among the objects. The problem was originally defined and a solution proposed by Bonchi et al. This paper discusses their work, drawbacks and proposes a modification to overcome them. The focus is to be able to identify larger groups from seemingly sparsely connected social networks.

The digital image processing is a fast pace growing field which requires pre-processing of the images before their actual usage and experimentation. Image Segmentation and Object Recognition is one of the phase which requires the prior information about the object to give the extracted attributes as output of the images. Here, in the present research work we have already some outdoor segmented images which have been identified as to which category of classification they belong. This work has been accelerated with the help of machine learning that is a novel technique to identify the objects with outstanding accuracy. The machine learning algorithms have aided in this process of identifying the objects like sky, brick, cement, grass etc. This collaboration of image segmentation with machine learning has proved to be accessible in large datasets where after segmentation images can classify themselves into a category provided it has attributes of the images.

Model checking is a very powerful formal verification technique. Formal verification of complex systems is a major challenge in many areas of human society. The verification of properties of these systems is recognized as a difficult problem and faces a number of practical and theoretical problems. The main limitation of the formal verification is known as the state explosion problem. In this paper, we discuss about two contributions to this problem for the improvement of performance in time and memory space.

Dynamic measures are the class of software measures which are obtained when the software is executing and hence can give accurate information regarding the run-time quality of software. For this reason, many dynamic measures have been proposed in recent past, however, little or no empirical evidence exists about the usefulness of these measures for software quality prediction. The objective of this paper is to empirically validate an OO dynamic complexity measure from authors’ previous work [19] to assess its ability for predicting maintainability as external software quality attribute. A controlled experiment is carried out in this regard and correlation and linear regression have been performed on the experimental data. The results of the experiment suggest that OO dynamic complexity measures can serve as a useful indicator of maintainability.

Cryptography is the first line of defense to protect the data from intruder. Symmetric cryptography and asymmetric cryptography are two cryptographic based algorithms that serve the security goals: confidentiality, availability, authentication and integrity. Asymmetric algorithms are known as public key cryptography that uses two keys: one public key for encryption and second private for decryption while symmetric algorithms are known as private key cryptography that uses the single private key for encryption and decryption. Symmetric algorithms are less costly compare to asymmetric algorithms. Normally asymmetric algorithms are used to distribute the secret sharing key of symmetric algorithm and symmetric algorithms are used for data encryption. Wide range of symmetric algorithms exists like Blowfish, DES, 3DES, AES, Twofish, RC2, RC5, CAST-128, and IDEA. Among these symmetric algorithms, AES and Blowfish give better throughput compare to other symmetric ciphers. In this article, we proposed the scheme to enhance the Blowfish block cipher security. In proposed scheme, total numbers of Blowfish rounds are altered by skipping few Blowfish rounds using round key. As a result, proposed scheme increase additional Blowfish cipher security against brute-force attack apart from minimum to maximum size of Blowfish key. In addition to that the proposed scheme also decreases encryption and decryption execution time of Blowfish cipher.

The current technological growth is tremendous so people are too much attached with technology. The most popular investing money portfolio is stock market and it’s too dynamic in nature so the risk is also high to loss the money. Now days lots of research ongoing to predict the price. This study considers news impact as semantic analysis and as a technical view stock prices and index is measured. The text mining is one of the latest technologies to perform textual based analysis. There are many techniques available to perform auto news classification which is one most important phase in this research work. This paper focus on comparative study about different available techniques for semantic analysis by measuring different parameters like the accuracy, the data set used. This paper concludes with best news classification approach for stock price prediction.

Requirements engineering (RE) is an important phase for data warehouse development. Earlier, DW development did not emphasize on RE phase. In recent past, several authors have suggested to give prime importance to this phase. Various approaches have been proposed for Data Warehouse Requirements Engineering which may be categorised as goal-driven, user-driven, mixed-driven etc. In this paper, we provide a detailed and updated review of goal oriented requirements engineering approaches proposed till date. It also provides a deep insight into goal oriented approaches through comparative analysis, which can be fruitful for enhancing the research in this field.

A naïve approach in rough computing under fuzziness and intuitionistic fuzziness through thresholds was given by G. Ganesan in 2005. Later in 2013, B. Krishnaveni and G. Ganesan had derived a procedure of characterizing information systems using intuitionistic fuzzy decision attributes through intuitionistic rough fuzzy groups. Using this, G. Ganesan and B. Krishnaveni in 2014 introduced the indexing procedure in characterization obtained using fuzzy decision attributes. In present paper, let us apply this procedure under intuitionistic fuzzy decision attributes.

A Wireless Ad-Hoc Network (WANET) is a group of wireless mobile nodes which are self creating, self organizing and self administering. In such networks, selection of routing protocols is an important concern as it affects the performance of ad-hoc networks in terms of average end-to-end delay, reduced packet processing etc. This paper proposes a Cluster Based Logical Hierarchical Addressing scheme for Dynamic Source Routing protocol (CBHDSR) which enhances DSR. This scheme reduces the flooding of RREQ packets in the network by searching a direct path based on the addressing scheme which results in the reduced congestion.

A design of an area efficient and low power 16 bit Multiply and Accumulate (MAC) unit is implemented in this paper. MAC unit performs various Digital Signal Processing applications generally contain number of repetitive methods having multiplications and additions. The MAC unit is designed by Modified Wallace Multiplier (MWM) using compressor with Carry Increment Adder and Carry Select Adder as final adder separately. The proposed design is implemented in Verilog Hardware Description Language (HDL) using Xilinx 14.5 Virtex7 and synthesis is done in Synopsys Design Compiler using Designware logic standard cell area library of 90 nm and 45 nm technology.

Since 1980s, Translation Memory (TM) have been accessible. It becomes an important language technology to assist the translation. It is a database that saves “segments”, which may be sentences, paragraphs or sentence-kind elements. Tree Adjoining Grammar (TAG) is planned to use along with Machine Translation System (MTS). To make efficient machine translation and to reduce the response time of online machine translation, we come up with the use of a TM. The combined architecture of machine translation with translation memory is indicated. To make the translator’s task faster, more efficient and easier, translator tools were designed. Translation tools were designed with the objective to minimize monotonous translation work.

In pervasive computing, human basic activity recognition has become one of the major challenges as recognizing every day basic activities and then classification of diverse activities using various devices has become arduous task. With the help of various machine learning models and data mining tools prediction has been applied. The dataset has total 10299 labelled activity instances with 561 features, to get the results more optimize we have successfully reduced features to 35 and the results were brilliant. Various machine learning models have been evaluated on the dataset for prediction of human basic activities. Results show that the large features and the reduced features were almost maintained in the terms of accuracy. The best models have been investigated for ensemble learning to get sustainable results on the basis of accuracy to classify the set of common activities carried on whole day. Encouraging results have been obtained with ensemble model. Cross validation has been performed to check the consistency of the ensemble model and accuracy more than 85% has been obtained. Finally, various human activities have been classified using ensemble model with good results.

Trends and technologies are changing very rapidly with time. In advent of 20th century, the scope of internet was limited and so forth the expectations of the users were also limited. Now a days; internet is in the reach of every one and therefore the way of use and expectations have changed a lot. Huge amount of data is being created every day. Social media has become very powerful. The volume of data is increasing exponentially every year. So it has become difficult to process huge amount of data to extract useful information. A new concept has been proposed to resolve the issue named as Big-Data. To use the concept of Big-Data, Apache has proposed a framework named as Hadoop. In this paper, a comparative study has been performed on Big-Data and serial Processing by using a simple word count example and found that the result obtained using Map Reduce (Hadoop) are encouraging as compared to traditional processing.

Smart meter is an essential component of smart grid that measures and transmits power consumption details of customer to control center for billing and monitoring purpose. The open nature of two way communication network used in smart meter is susceptible to cyber attacks. Hence the cybersecurity gets high priority in smart meter applications. Key management system is an essential requirement of cyber security. Implementing it to dynamic, large scale smart meters by existing methods is more complex and also it is not possible to authenticate the key derived from genuine user. The emerging bio-cryptography provides solution to these problems by generating keys from user biometrics in smart meters. In this paper, an efficient key management system is proposed in which keys are derived from a bio-cryptography technique using fingerprint biometrics with Euclidean distance algorithm. The proposed technique is simulated in Matlab and its performance is evaluated. Simulation results demonstrate that the proposed technique is well suitable for data security in smart meter applications.

Vehicular Ad-Hoc Network (VANET) has gained the interest of the researchers in the current decade. VANET can be implemented to improve safety of vehicles on road and offer greater comfort to the drivers and passengers in traffic. Due to high mobility of vehicles frequent handoffs are required in VANET. These frequent handoffs must be carried out in an efficient manner with reduced handoff latency. To reduce the handoff latency during a switch, the scanning time to search for the new access point must be reduced significantly. Various schemes have been proposed which perform a pre-scanning in order to reduce handoff delay. This paper presents a survey of such pre-scanning schemes and tries to explore new areas of enhancements which will result in decrease of handoff latency. A qualitative comparison of these schemes is also carried out which leads to newer areas of research, where significant improvement can be beneficial.

An adequacy criterion in software testing are the rules or guidelines for quantitative analysis of any test cases and sets the target to be achieved as to stop the testing process or test data generation. Search based test data generation is the application of the Evolutionary Algorithms to achieve an optimized test suit which is tested against the fitness function to check its distance from the target. The Entropy of Information Theory can be defined and the measure of “uncertainty”. The paper is the study of the relation between the entropy, Information Gain and the uncertainty regarding the random generation of test data.

Electricity bill generation and meter reading is a very complex process in today scenario. The existing method of billing process uses manual work of taking reading of the meter, updating details of meter and sending bill to the customer. Till now, this is being implemented by using OCR but it does not allow user to track the consumption each second. We are developing a method on IoT platform which is more efficient and can be used for keeping track of electricity consumption. User can get the reading on android any time. Our electricity distributor will no longer need to take reading manually. The data will be directly transferred to them which will be used to create electricity bills. For multiple meters in single household, combined result can be viewed. Also, Client is able to view the graphical usage for a particular duration.

A new single layer simple probe feed asymmetrical koch fractal dual band microstrip patch antenna (MSA) is proposed for ISM band (2.4 and 5 GHz) applications. To achieve circular polarization (CP) and dual band resonance, koch fractal curves are embedded on sides of square patch. The radiating cross bisected antenna resonates at ISM band. Without increasing structural complexity, improvement in bandwidth is achieved. Proposed design fulfils requirements of IEEE 802.11a/b/g standards. The measured 3-dB axial ratio bandwidth is 3.48% (2.4–2.485 GHz) and 12.31% (5.15–5.825 GHz) and 2-dB VSWR bandwidth is 6.1% (2.35–2.5 GHz) and 15.55% (5.10–5.96 GHz) respectively at center frequency of 2.45 GHz and 5.5 GHz. Experimental results show perfect match between simulated and measured bandwidths.

This paper provides a comparative analysis of various graph based extraction methods for automatic text summarizations using ROUGE on review dataset. We consider five techniques that include TextRank, LexRank, LSA, Luhn, and Edmundson. These methods concentrate on predicting the semantics of an entity. The experimental results on summarizing the users’ opinions show that the LexRank method gives the best performance among all. Generated summaries are understandable and convey informative opinions.

Hybrid watermarking techniques are gaining importance in the field of image watermarking as they meet criterions like Imperceptibility and Robustness. A novel approach for the hybrid color image watermarking using Cycle Spinning based Sharp Frequency Localized Contourlet Transform (CS-SFLCT) and Principal Component Analysis is proposed. Contourlet Transform (CT) is applied to decompose the images into various sub bands for all the color spaces of both the host and water mark images. For inclusion operation principal components of middle band(X bands) are considered and CS-SFLCT is chosen for watermarking. CS-SFLCT provides better frequency localization resulting in a very good PSNR and Principal Component’s helps in fruitful extraction of watermark from the host image. This hybrid technique has shown better correlation between input watermark and extracted one and shown very high very high robustness under various intentional and non intentional attacks.

The data usage pattern is changing rapidly in many real life applications and these applications have converged in Smart Phones. 5th generation wireless networks will envision a widespread use of Internet of Things (IoT). With the growing demands for communication between Internet of Things using 5G networks, securing devices will emerge as a big challenge. There will be hidden exchange of data between the devices for which security can be achieved with audio steganography. The aim of the paper is to formulate the model supports the methodology and infrastructure desired to implement the security for IoT in 5G networks. Audio steganography is an invisible communication used for hidden exchange of the data. Internet of things applications varies from ubiquitous computing to machine to machine communications with most applications will be on Voice over IP and will require securing of data from eavesdroppers and attackers. So, the focus of this research paper is on how can we secure internet of things using audio steganography in 5G Platform.

Software advancement has increased the complexities many fold and to meet the quality standards, a lot of research is being done in designing new testing methodologies and tools. Mutation testing is a proven effective technique but the high cost attached with it averts it from establishing it as an industrial tool. The review is an extension of the previous work where a review was done on search based test data generation and mutation testing. The objective is to study the remaining techniques/approaches and summaries the discussion of both the reviews. The application of mutation testing with various techniques at various phases of software development along with various languages/tools show that it is a versatile, adaptable and efficient, which is motivating the researchers to explore the new areas.

With exponentially increasing users and crucial information accumulation over the internet, it has become a necessity to introduce a system which is powerful in terms of providing protection, and effective in cost required in authentication process. Biometrics is the only thing that cannot be stolen or copied as every human being has their own unique features that cannot be imitated by any intruder. The only disadvantage of biometrics authentication process, the need of additional devices is removed in the proposed system. This method can make the computer uniquely identify a user by typing behavior and defeat intruders. As accurate as any other biometric security technique, keystroke biometrics is cost effective because it does not require any additional hardware. In this paper a new system is introduced having keystrokes biometric added with password hardening techniques as an effective authentication method to defeat intrusion attempts.

Cellular concept is major breakthrough in solving spectral congestion but it faces spectrum availability crisis due to increase in requirement of data rate. In order to satisfy the growing traffic demand and to optimize channel allocation, cognitive cellular network (CCN) is taken into consideration. CCN consists of primary (cellular) and secondary (cognitive) users where secondary users try to accommodate in primary band without causing interference to them. Artificial Bee Colony (ABC) and Teaching Learning Based Optimization (TLBO) algorithms are the population based optimization techniques, taken into account to analyze and improvise the network capacity. In this paper, these techniques are studied and used to find SINR in terms of data rate in Mbps with the help of fitness function. The obtained results of ABC and TLBO are compared with reported work of Particle Swarm Optimization (PSO). The proposed techniques are found to be better than PSO.

The paper explains an optimal design of fractal antenna using modified Sierpinski Carpet geometry for wireless applications. The proposed antenna is designed on substrate (FR4 glass epoxy) by considering the thickness of 1.6 mm and Ɛ_r = 4.4. The resonant frequency taken for proposed antenna is 2 GHz. It is observed that on increasing the antenna iterations the gain also increases with it. The (HFSS V13) High Frequency Structure Simulator is used for designing and simulation of proposed antenna. The performance parameters of antenna like Voltage Standing Wave Ratio (VSWR), Return loss and gain for different iterations are also observed and explained in this paper.

Performances of current vehicle supervision techniques are require being improved using portable sensing scheme. In this paper the technique used is advance projected system for parameter measurements of vehicles using AMR sensors. Objective of the projected system is to develop an automated and improvised system where there are 3 AMR sensors that give height of vehicle and speed estimation and number of vehicle passing near range of sensors scheme. Thus, speed, classification and total count of vehicle these parameters are expected correctly.The anisotropic magnetic sensor scheme rooted alongside the road-way and tested the traffic measurements. Calculation of Td (time-delay) and distance among two sensors estimate the speed of vehicle. This system is suitable for reliable count of number of vehicles. Height of vehicle is calculated for the categorization in low and high level vehicle. This system is simple and utilize in real time measurements. The system gives the average speed of total low/high vehicle pass from the track of sensor scheme. The result gives improved presentation for constraint measurements.

Fast growing internet coupled with advancements in image processing technology, has resulted in increased incidents of image deception. Digital image watermarking is widely used as a tool to establish ownership and restore the trust in digital images. The paper presents a study of solutions as proposed by Ali and Ahn in [1] to overcome the problem of false positive detection by SVD based technique, specifically the one proposed by Agarwal et al. in [2] The solutions are tested against re-watermarking as another class of ambiguity attacks.

Clustering is done in order to group the entities which are alike in one group so that grouping of more similar objects can be done. The objects placed in one group are known as clusters. In this paper we are using clustering in order to identify the human locomotion and categories the dataset making clusters. We are using two clustering techniques i.e. SOM and K-mean. So we first selected the feature and identify the principle feature then we cluster gait data and use different machine learning technique (K-mean and SOM) and performance comparison is shown. Experimental result on real time datasets propose method is better than previous method as far as humanoid locomotion classification is concerned.

Images are normally degraded with noise. The main goal of the denoising technique is to eliminate the noise with minimum distortion. In this paper, work has been done to remove the salt & pepper noise from some of the standard images. The image denoising has been performed with median filter (MF), adaptive median filter (AMF), decision based unsymmetrical trimmed median filter (DBUTMF), modified decision based unsymmetric trimmed median filter (MDBUTMF) and decision based unsymmetric trimmed midpoint filter (DBUTMPF). The performance of each technique has been evaluated on the basis of four parameters namely, signal to noise ratio (SNR), Structure similarity index measure (SSIM), edge preservation index (EPI) and multiscale structure similarity index measure (MSSSIM).

Text summarization is a method which generates a shorter and a preciseform of one or more text documents. Automatic text summarization plays an essential role in finding information from large text corpus or an internet. What had actually started as a single document Text Summarization has now evolved and developed into generating multi-document summarization. There are a number of approaches to multi-document summarization such as Graph, Cluster, Term-Frequency, Latent Semantic Analysis (LSA) based etc. In this paper we have started with introduction of multi-document summarization and then have further discussed comparison and analysis of various approaches which comes under the multi-document summarization. The paper also contains details about the benefits and problems in the existing methods. This would especially be helpful for researchers working in this field of text data mining. By using this data, researchers can build new or mixed based approaches for multidocument summarization.

Wireless Ad hoc networks represent a form of cooperative networking through peer to peer behavior with others nodes in the networks. Hop by hop communication is default way of communication. Most of the communications are localized and interaction among local nodes requires local security provisioning. In the absence of any centralized certification authority and absence of viable localization and synchronization hardware, schematic localization and periodic refreshing proved to be a feasible solution. Several solutions have exploited GPS based localization and periodic refreshing cycles to provide a viable security solution for wireless ad hoc networks. In this paper, we have proposed an accelerated hashing mechanism with schematic localization based on variable or multiple transmission range of few nodes. The solution has been simulated for evaluated for performance parameters like connectivity, storage overhead and computation.

In this paper, a novel dynamic routing protocol is proposed so it is suitable for port addition, removal and link/port failures. The proposed protocol is favorable for virtual cellular networks and it does not suffer in the transmit power when it needs to rerouting phase. The transmit power of the proposed rerouting is compared with that of the complete power minimized one. The simulation results shows the transmit power and time and number of messages for route reconstruction in the paper protocol is efficient.

We propose an innovative approach towards Cooperation Models for Multi-agent Reinforcement Learning (CMMARL) using reinforcement learning methods. Communication methods for reinforcement learning depend on multiagent scheme is proposed & implemented. Different cooperation methods for cooperative reinforcement learning based on expertness measure of each agent proposed here i.e. group method, dynamic method, goal-oriented method and expert agent method. Implementation results have demonstrated that the suggested communication and cooperation methods are able to accelerate the aggregation of the agents that accomplish best action strategies. This approach is developed for dynamic products availability in a three retailer shops in the market. Retailers can cooperate with each other and can get benefit from cooperative information by their own policies that accurately represent their goals and interests. The retailers are the learning agents in the problem and apply reinforcement learning to learn cooperatively from the situation. By making considerable theory on the dealer’s inventory strategy, refill period, and entry procedure of the customers, the problem turn out to be Markov decision process model thus facilitating to apply learning algorithms.

Nutrition and obesity related diseases are now become dangerous to human health. Now a days people go on dieting and not taking proper care of their health and food intake calories. So to solve these problems a system is developed which is automatic dietary food intake detection system by using wearable sensors. It consists of embedded system and signal processing system in which food intake is sensed by high fidelity microphone and signal is pre-processed by embedded hardware part and through bluetooth it is sent to smart phone. Here we have used hidden markov models to recognize chewing and swallowing events to extract time and frequency domain features as well as volume and weight of food intake. Algorithm for decision is developed for types of food detection. For e.g. in diet control patient suffering from diabetes precisely monitor daily food intake. An application on smart phone is developed to show the results of food intake as well as give guidance for better eating habits.

Organizations have more data than they can use effectively. Organizational leaders are now aware of the need to employ analytics to exploit their growing data and use computational power to become smart and innovative. Senior management now a days demand that businesses should run on data driven decisions and expects data to be churned out into information very quickly. Data analysis is required for making real-time decisions. Better reporting and decision making enables the industry people to excel in the areas of sales, marketing, delivery and operations [11]. Companies are heavily investing in technologies that help them collect data, which are mostly unstructured and semi structured, analyze it in real time to understand consumer’s sentiments and take appropriate actions. This paper is an overview emphasizing the need of data analytics in business.

In this paper a multi-sensor data fusion approach for wireless sensor system based on Bayesian strategies and Ant Colony Optimization procedures have been recommended. In this methodology, every node is furnished with various sensors (i.e. temperature and humidity sensors). Usage of more than one sensor gives additional information about the environmental conditions. The data fusion based on the competitive type hierarchical processing has been considered for experimentation. In the data fusion, the data gathered by the sensors are set in the sensing fields and afterward the data fusion probabilities are registered. In our suggested approach, the gathered temperature and humidity information are prepared by multi-sensor data fusion strategies, which then helps in diminishing the energy utilization and also communication cost through accumulation of the repetitive information. The multiple information merging process is reliable and accurate in addition to being energy-efficient which our primary goal is. The proposed algorithms were simulated utilizing Matlab. The implementation of the proposed algorithms were conducted with and without multi-sensor data fusion and the outcomes demonstrate that the proposed algorithms could reduce the energy- consumption, rather save additional energy, thus enhance the entire lifetime of the system.

Frequent pattern mining is top chart research field for young researchers. It has a huge array of real world applications. Although many algorithms, tools, techniques are available for performing the task of frequent pattern mining. Apriori and fp growth are very popular frequent pattern mining techniques. This paper presents an updated methodology for frequent closed item set mining. The proposed model is based on the concept of data reduction. Useless data is eliminated from the transaction data base. The experimental results have shown that the proposed updated method is outperforming the existing methods.

EUTRAN LTE–A have the capability for adaptive connectivity between base stations (eNodeB or eNB) and its consistent remote radio head (RRH)/relay node (RN). In this paper, we have described the effect of downlink multipoint connectivity with interference. We analyze the performance of Transmission Point under imperfect CSI with Coordinate Multi Point (CoMP) connectivity. In the simulation results, we found Multi Point connectivity enhances throughput with more reliable connectivity of UE with core nodes. Apart from these we measure performance of direct path (UE-eNB) and alternate path (UE-RRH/RN-eNB), and found an improved radio signal in a relationship of throughput, Bit Error Rate (BER), and Block Error Rate (BLER). Also observe throughput performance of the direct path is higher than alternate path due to transmission inherent effect.

Malware is an abbreviated term meaning “malicious software”. This software has a capability to gain access or infect a system without the knowledge of the owner. In this paper, we have tried to provide brief information about different types of malwares known till date such as virus, rabbits, botnet, adware etc. Apart from those we have mentioned the cure to it i.e. intrusion detection. We have described various techniques of intrusion detection such as signature based, anomaly based, behavior based etc. Methods for implementing these techniques include neural networks, data mining etc. A brief description of Intrusion detection system is also provided which is a software application used to monitor the network and system activities and also to detect malicious actions. The objective of this paper is to provide complete study about the types of malware, techniques and methods of intrusion detection, challenges and applications.

This paper deals with Drop size distributions and number of drops and its volume in a frame. We are using video to frame conversion so that we can calculate the amount of rain water in a particular time. This system found to be more accurate and errors free like spreading, evaporation of drops induced errors are neglected as this system uses image processing tool to make analysis of data. It involves many operations which performed on raw data collected from high definition camera. System response time is also less.

IEEE Standard 1232 provides the D-matrix for diagnosing quality in models. The framework give the ability to detect dependency in relation to symptoms and failure modes [1]. This paper describes an approach to construct D-matrix by mining unstructured repair verbatim text. At first d-matrix is constructed for different dataset, and then we can form a combined d-matrix from different dataset to identify common patterns in it. In this proposed method training is performed by using different classification methods on unstructured verbatim (Combined D-Matrix) collected from the medical domain.

Wireless sensor network is of prime importance because of its applicability in various domains ranging from healthcare applications to military applications. Security of such networks is important as these carry confidential information. Security of Wireless Sensor Network is divided in three phases, prevention, detection and mitigation. In Prevention phase, care is taken so that attack should not occur. But most of the times it is not possible to prevent attacks, so it is very important to detect them as early as possible so that these will not harm a lot to wireless sensor network and that phase is called as intrusion detection. Once Intrusion has been detected we have to take actions to cure from it and it is called as mitigation. So, Intrusion detection is most important phase as far as security of wireless sensor network is concerned. This paper discusses about various detection methodologies such as Anomaly based, Misuse based and Specification based IDS, various decision making schemes for intrusion. Major focus of this paper is to understand various Intrusion Detection Systems that are proposed for wireless sensor network. These are discussed with different issues with advantages and disadvantages. Finally paper gives future directions for selection of Intrusion Detection System.

Glacier mapping accuracy plays very important role in studies like mass balance of glacier, water resource management and in understanding the health of the glacier. Several of the present glaciers are covered with debris of different thickness. So it becomes difficult to distinguish debris covered glacier from the adjacent valley rock, alone with the use of optical data because of the same reflectance in visible to near infrared region. In this paper we have trained Support vector machine (SVM) and Artificial neural network (ANN) on several parameters such as slope, surface curvature, thermal data and also on several texture parameter, such as variance, skewness, entropy, homogeneity, mean and dissimilarity. Then both the algorithms were applied on the part of the alaknanda basin. It was observed that both ANN and SVM produced good results, with accuracy of SVM slightly higher than that of ANN algorithm.

Face recognition from surveillance camera is a challenging task due to variation in lighting conditions, motion blur and poses. Most of the face recognition algorithms perform well under controlled environments. In uncontrolled scenarios, face recognition algorithms are being developed to operate on information fused from multiple cameras. This approach increases the hardware and processing speed. In this paper effect of concatenating multiple features on the face recognition rate is being investigated. The developed algorithm is tested on the publicly available chokepoint dataset. Recognition rates achieved by concatenating multiple features are found to outperform the results of the methods using information from multiple cameras for face recognition. Further testing with various features need to be performed.

MANET is a set of mobile nodes in which communication occurs between them using wireless links. Infrastructure less, dynamical topology and Lack of central communication of nodes makes it vulnerable to various kinds of attacks. One of the major security problems is Black Hole attack in which node silently drops the packets in the network. In this paper, we propose a solution to mitigate this attack in MANET using mobile trust points with clustering. The proposed method uses some mobile trust points which monitor the activities of cluster heads to detect the attack and then generate alert in the network if any black hole node detected.

Watermarking methods have been utilized for safeguarding contents against unlawful replication. New immune watermark introduction and retrieval mechanism constituting FFT based approach along with chaotic encryption is presented here. Encrypting the procedure develops a secure environment for the scheme. The primary audio undergoes Fast Fourier Transformation. Introduction of encrypted secret contents is accomplished adopting Fibonacci numbers. Retrieval of mark is achieved by non-informed scheme. Error rate as well as SNR outcomes are the execution specifications computed for this method. Here robust quality in opposition to distinct processing actions like reverberation, echo and smoothness is administered. Also greater imperceptibility procured by this method demonstrates the proprietary rights of initial content.

The enormous applications of kinematic chains (Mechanisms) has lead to more scope in the structural synthesis of the same. The interdisciplinary applications like in mechatronics, dental, medical (especially in human bone fractures), ergonomic design of machinery, physiotherapy machines, industrial robots, increased use of robots in every field, atomization in the industry etc. in almost every field the kinematic chain, being the basic element is necessary. Hence the more concentration is given to enumerate the feasible kinematic chains. The coding (Maxcode) methodology adopted for the enumeration of feasible, nonisomorphic, distinct kinematic chains has given an added advantage in the enumeration process by eliminating the infeasible, isomorphic chains during the process of enumeration. The present paper presents the methodology and the results of enumeration of nine link two degree of freedom kinematic chains.

As the Internet is expanding, the threat of intrusion is also increasing. Modern businesses which use Internet demand for strong computer and network security. Intrusion prevention is obviously the best choice from security viewpoint, but it has practical limitations as hackers develop new methods to breach security. Thus early detection of an intrusion is the sensible option and Network Intrusion Detection Systems (NIDS) carry out that task. In this paper, the authors propose extended empirical evaluation model specifically from NIDS perspective. The objective is to introduce multiple classifiers into the model along with feature selection method for improving performance of the classifiers. Additionally, the new model incorporates a feedback mechanism to ensure new prediction learn from rectifications of past records.

The data centric networking resources are virtualized with the concept of Software Defined Networking (SDN). This is used to overcome the traditional networking issues such as traffic congestion, traffic delay, and to improve the network scalability. The networking topology is emulated in the mininet emulator. The customized traffic policy is implemented in the SDN controller. In this paper, mininet emulator and the OpenDaylight SDN controller are studied thoroughly in order to establish the SDN environment. The research challenges are discussed to set up the virtual network in the cloud based data center.

With advanced telecommunication systems, need for antennas with multiband features for wireless applications are increasing. The paper presents hexagon shaped microstrip antenna operating at 2.47 GHz with gain of 7.96 dB. The hexagon shaped antenna is converted to multiband antenna by enforcing fractal concept in the geometry of antenna. The iterations of the antenna are done in such a way that antenna size is reduced by 49% as compared to base geometry conventional antenna. The performance of antenna is evaluated using HFSS (High Frequency Structural Simulator); v13 software from Ansoft. Results show that improved performance parameters for instance gain, return loss, VSWR, bandwidth etc. are achieved with co-axial feed. The closing iteration is designed by FR4_Epoxy material having dimensions 27.5 (L) × 36.5 (W) × 3.2 (h) mm3 with εr = 4.4 appropriate for S-band applications.

The purpose of automation should not be restricted to power conservation but it should also facilitate ease of control of devices with minimal intervention of human being [1]. In this paper we have proposed an alternative approach to allow the human to control the home appliances through their normal blink of human eyes with occurrences of reflected infrared light by the eyes pupil. In this paper an IR emitter and detector aligned at such a coordinates that the infrared light would be able to get reflected through the pupil. It is connected to Universal Board by wireless connection which is being used to control the supply of current to the appliances by a two way switch connection. An Infrared Emitter is mounted over the main device frame to pin point the particular device. The main eye device has a timer switch which turns OFF the device after few seconds for minimum exposure of IR on eye and push button to turn it ON.

Social network mining is a growing research area which combines together different fields such as machine learning, graph theory, parallel algorithms, data mining, optimization, etc., with the aim of dealing with issues like behavior analysis, finding interacting groups, finding influencers, information diffusion, etc. in a social network. This paper deals with one of these important issues i.e., Influencer Identification in social networks. This paper presents a data mining modelling approach for a twitter network, to find the most influential user among the given pair of users. This could be scaled over the entire network. We used a data mining model to score the test data and predict the influential user among the given pair of users. This approach of modeling can potentially be used for building many of the marketing and sales strategies wherein the influencer may be motivated for diffusing information or new ideas.

Placement of artificial magnetic conductor (AMC) structures at the inset feed line of the patch operating at 2.4 GHz is analyzed. Proposed design for four element patch array using AMC with multilayer PCB technology which gives improvement in return loss and gain. Analysis had done for low cost stack up material for multilayer purpose. FR4 epoxy finally used as stack up material which is low cost and ease to fabricate. In antenna array AMC structure at the feed line considerably reduce side lobe level and back lobe level and improves gain. Simulated results are validated experimentally.

Vertical handover is one of the most prominent challenges in heterogeneous networks (Hetnets) since most of the user devices come with mobility feature. In order to provide a seamless handover between various network topologies, several additional parameters other than signal strength must be taken in account to satisfy user preferences at an acceptable level. This paper proposes a fuzzy logic based multi-input criterion for handover decision in wireless heterogeneous networks that uses received signal strength indicator (RSSI), monetary cost, data rate and mobile station (MS) velocity as the input parameters. The simulation results show that proposed fuzzy logic based algorithm gives an improvement in reduction percentage of number of handover as compared to existing systems.

There is a big-data explosion in the field of bioinformatics, with the rapid growth in the size of biological data. In bioinformatics, workflows are used to integrate and analyze biological data. Orchestration and choreography are the two approaches used to execute bioinformatics workflows. However, big-data poses several challenges in these approaches. One of the challenges is how to handle the movement of big-data during workflow execution. With the advent of big-data, a number of modified orchestration and choreography approaches have also been developed to handle big-data. In this paper, we review and make a comparative assessment of the state-of-the-art approaches to execute big-data workflows. We examine the big-data handling in these approaches and finally recommend a solution that could be a way forward in executing big-data bioinformatics workflows.

In this paper, we have discussed several code replication detection methods and tools in different dimensions. This review has provided an extensive survey codec clone detection techniques and tools. Starting from clone perceptions, classification of clones and an overall assortment of selected techniques and tools is discussed. This paper covers the whole paradigm in clone detection and presents open research avenues in code clone detection.

Steganography is a technique of hiding the private messages inside the cover image. The main objective of steganography is to send information or message in hidden manner from source to destination in such a way that the intruder or attacker cannot crack the contents of the message and even would be unable to feel the presence of secret message. The proposed method introduced as a new hybrid security model based on steganography. AES algorithm with chaos function to encrypt secret message on first level has been used and RSA algorithm to encrypt secret message on second level have been used. LSB technique is used to hide encrypted message into cover medium. The proposed technique is tested on various images. The PSNR value is calculated for better picture quality of stego image.

The rapid growth in population has led to the increase in the amount of solid waste disposed at dump yards. Every year, several tons of solid waste is being deposited into the Pallikaranai marshland, Chennai, India impacting the environment ominously and causing serious health issues. Using time series satellite images and gas sensors, the environmental changes over the years are determined and environmental changes in the near future are predicted. Satellite images of the area under study are obtained chronologically from CNES/Astrium and DigitalGlobe. The approximate area of the dump yard in each image is calculated by using Google Maps API. The variation in the area occupied by the dumpyard over the years is determined by using change detection technique. Gas sensors are used to measure the levels of various harmful gases such as methane compounds, Carbon Monoxide and Carbon Dioxide. Air Quality at the dump yard and a couple of locations near the dump yard also determined. Based on the results, the possible environmental changes in the near future are predicted.

Developing quality software is one of the most challenging tasks, for developing quality software we have to remove the entire bug from software before the software switch into operational phase. For this we have to allocate our testing and debugging resource based on the time so that we can finish off our work. For distributing the testing and debugging resource we are using software reliability growth model (SRGMs). Numerous SRGMs has been developed in past couple of decade for allocating the testing and debugging resource but mostly under static condition. In this article we developed a mathematical model for allocating the resource in dynamic environment. In this article we utilized Pontryagin maximum principle for illuminating the model. Finally one numerical illustration is explained for distributing the software testing resource for created module. Here Genetic Algorithm (GA) is used for allocating resource optimally.

This decade has seen massive explosion of information and data, voluminous data bring about many challenges – data storage, data analysis, noise accumulation, data privacy, security and heterogeneity of data. With the emergence of this voluminous data also known as Big data, an explosion of data have been seen in all fields related to science and engineering. In current scenario this term often applies to data sets with extreme size (Exabyte’s). In this article we present an overview of the need of big data, its applications and challenges.

In recent studies video surveillance become an important task for identifying the motion detection (moving objects) to avoid the burglars in home security systems. Most of the video surveillance systems works on algorithms like background image subtraction, double background filter (DBF), optical flow method for motion detection where video is recorded by digital video recorder when an moving object is identified to save the memory, but in due process some frames are missed even moving objects are identified as they are treated as stable objects (without considering the minor movement in objects under threshold value), here we propose a method to integrate background image subtraction and double background filtering with morphological dilations to identify the moving objects and enhance the video quality by using the trained filters to improve the low quality frames.

The work described in this paper intends to explore the possibility of using electronic stethoscope as potential analysis tool to aid automatic diagnosis of heart related problems/diseases. Heart sounds plays an important role in preliminary diagnosis of various cardiac disorders. It is pertinent to check the fidelity of electronic stethoscope for its usage as primary auscultation device in telemedicine applications. 3M Littmann 3200 electronic stethoscope was chosen for experimentation due to its popularity amongst medical practitioners. The features and operating parameters of the stethoscope are analyzed and frequency response in different operating modes is observed. The various aspects of stethoscope in phonocardiology are discussed with reference to future scope of implementation.

Software plays a crucial role in day to day life; hence its security and reliability cannot be neglected. Creating a secure software system is not just to secure sensitive and confidential information but it needed to establish a system which could stand true on the benchmark set for being a secure software system and further derive a roadmap to construct impregnable and efficient software. In order to fulfill this criterion, security testing is vital for the development of a secure software system as it pursue all the aspects of SDLC. Security should form an integral part of a SDLC, hence to maximize and maintain the defenses of a software system and to keep its development cost in limits, Security Testing Profile (STP) provides a reliable platform for testing software. STP is an uncharted territory and more progress can be made in this area, which may help in developing robust software systems.

Secret communication in Wireless Sensor Network is considered to be a challenging task due to high risk of node capture. Secret communication is only possible when communicating nodes share a common key for cryptographic purpose. Sensor nodes are often loaded with required secret key in its memory prior to deployment called key pre-distribution scheme (KPS). Various KPS have been proposed to provide secure communication and to minimize damage caused by intruder. In this paper, we will discuss properties of expander graph that are suitable in designing KPS and perform various experiments on KPS based on Ramanujan Expander graph for performance evaluation.

The Internet is becoming part and parcel of everyone in everyday life. Almost all people are depending heavily on the Internet for all types of online activities. Hence, many Information Technology (IT) companies and organizations do business with people or enable businesses for people by running and/or supporting online web services continuously and try to attain or guarantee continuous service availability. From the old and recent bitter experiences of web servers with Distributed DoS (DDoS) attacks, one can realize that they become a serious threat to web site’s availability. Along similar lines, flash crowds also prove to be damaging to web servers by losing business, if not properly discriminated from DDoS attacks and handled well. In this context, the task of differentiating the DDoS attacks from the flash crowds gets more significance and hence, we carried out an extensive survey over this. We analysed numerous existing approaches and compared important features used for discrimination.

In this paper a survey on design options is carried out. An investigation of comparative analysis of 1-bit full adder with various options on the basis of power dissipation, propagation delay, and area concerning to the of number of transistors used, Energy Delay Product (EDP) & Power Delay Product (PDP).In present scenario, reversible logic based designing is attracting researchers because of its less power usage. Reversible logic is playing important role in low-power circuit scheme. The application areas of reversible logic is nanotechnology, CMOS(less power), cryptography, DSP, DNA & quantum computing, communication, computer graphics. To contract quantity price, the profundity of circuits & count of drivel outputs are the main aim of intending reversible logics gates. This work will give brief idea about building of full adder circuits using the basic reversible gates.

Achieving security is a key aspect for any computer system. Many modern technologies have been applied to achieve the required security. Cryptography provides a primary way to achieve best security. A recent trend shows that many of the cryptographic algorithms are modified with new functionalities to provide better security in all aspects. One major research branch of Cryptography is Public key cryptography. In this paper, one of the popular public key cryptography algorithms, RSA with arithmetic functions are reviewed and analyzed. This paper mainly focused on the use of Carmichael function instead of Euler totient function applied on RSA algorithm. Results have shown that use of Carmichael function results in smaller value for decryption key. This leads to reduced decryption time of RSA algorithm.

In many application areas like video scrutiny, biomedical scrutiny the human motion detection from videotape is the point of interest. This paper represents the progression in topic to epilepsy, in which human motion is most vital element of patient’s clinical video. This paper illustrates the topical achievements in video processing, scrutiny along with identification of human motion in epilepsy designed for marker free system. Epilepsy is a disorder of CNS characterized by loss of cognizance with paroxysm. Seizure defines a sudden occurrence of disease. These seizures are attended through hysterical, frequently regular actions of body parts when seizure activity starts, brain areas amenable meant for unveiling with restraint of movement. The dynamics of these modulations is usually indefinite. In order to obtain adequate data for verdict and to plan remedial strategy human have to be monitored for long duration. The primary principle of the manuscript is to present a method by which clonic as well as tonic seizures can be detected using video processing. The proposed algorithm specifies optical flow for motion detection; global group transformation velocities for feature extraction with band pass temporal filtering to classify the incidence of tonic and clonic movement in video string. This paper shows a substantiation set of 10 prerecorded epileptic seizures, proposed system is extremely sensitive and precise in detecting recorded string containing tonic and clonic actions.

Diabetic retinopathy is an eye disease due to diabetes, which is not detected in its early stage, may cause vision loss. The need of automated diagnosis methods of diabetic retinopathy increases day by day because of its severity. Authors proposed the design of diabetic retinopathy automated diagnosis system based on neural networks. Multi Layer Perceptron (MLP), Principal Component Analysis (PCA), Generalized Feed Forward (GFF) neural networks are employed to design automated classifier system in first experiment. In second experiment, the input dimensionality reduction method based MLP, GFF neural networks classifier systems are designed and compared the performances. In experiment 1, the average classification accuracy for MLP network is nearly 99.00% whereas GFF-NN has 92.00% on CV data. In experiment 2, using Principal Components (PCs), the average classification accuracy for MLP network is nearly 97.22% whereas GFF-NN has 84.37% on CV data. The N/P ratio for MLP and GFF networks is large in second experiment which is 0.273 and 0.219 respectively having less neural network’s architecture complexity.

Nowadays Capacitive Micromachined Ultrasonic Transducers (CMUTs) appeared as a preference over piezoelectric transducers in terms of bandwidth, fabrication of layer arrays, efficiency and sensitivity. This paper presents the CMUTs cavity filled with air with different membrane materials namely silicon, silicon nitride and polysilicon. The operation of the device is discussed in detail with various electromechanical parameters like pull in voltage, Eigen frequency and deflection of membrane with applied DC and AC bias along with comparison between square and circular geometries of CMUT. 3D analysis is carried out in COMSOL where Solid Mechanics, Electrostatics and Moving Mesh modules have been combined to model the dynamics of CMUT. Finally, the time evolution of the device is derived for several frequencies, from where the maximum frequency of operation is obtained and comparison is demonstrated to help the researchers.

This is a major concern for most of the countries all over the world to fight against terrorism. Post 9/11 attacks it became important for the legal enforcement agencies to intercept such attacks in advance, otherwise at least identify the key suspects and their associated network. Although the huge amount of information generated and lack of effective analysis approach and methods impede the effective analysis. It cannot be done on the basis of traditional approaches and defend criminal and terrorist activities. However a novice approach like Investigative Data Mining (IDM), a type of Data Mining may provide a well defined approach to tackle the uprising situation. In this paper, an effort is made to understand and illustrate how IDM works and show its importance in identifying key nodes in terrorist networks with social network Analysis.

Android is an open source operating system with equal proportions of pros and cons. Android provides third party applications that include access to network communications, personal information, storage, location etc. The diverse range of applications available makes life easy, but comes with associated risks and potential threats to the user’s privacy. Each time a user installs an application, they are presented with access permissions required to install the application. Users are generally not able to understand the aforementioned permissions due to highly technical jargon presented. In this paper we are focusing on designing an application having access to all the modules (resources) in the smart phone. This application will let the user to select the modules which user wants to be secured. Once this application is installed, user will select few modules to keep private, for example let us consider camera, gallery, etc. Once selection is done, this security application will be running always in the back-end to monitor the security for the modules selected. If Any other application tries to access the secured module, user will get the notification saying that active app is trying to access the camera and needs permission for its working. User can select whether or not to grant the access and also save the selected preference for future.

Quality inspection of fruits with help of computer vision or image processing is gaining much attention nowadays because of the costly and labor intensive techniques used earlier. It has been proven much useful to the agricultural sectors and other industries in the past. The present research work is based on the quality evaluation of apple fruit. Since less work is done for the quality evaluation of internal part of apple, i.e. slices, here the experiment will be done; first on the external surface and then on the slices. Here hue histogram intersection is used for external surface defect detection and for slices defect checking; features called Color Coherence Vector and Complete Local Binary Patterns are extracted from the slice images and they are given as input to Multi-Class Support Vector Machine classifier. Both linear and nonlinear SVMs were used and linear classification gave better results.

Protein function prediction is a challenging classification problem. A computational method is vital to perform the function prediction of Proteins. For this various Feature Selection techniques had proposed by eminent researcher. But several techniques are model based or for a specific type of problem. In this paper, we make a comparative analysis of different supervised machine learning methods for the prediction of functional classes of proteins using a set of physiochemical features. For an attribute or feature selection we have used a novel hybrid feature selection technique to overcome some of the limitations of existing technique and also present a comparative analysis of the classification of enzymes function or family using different computational intelligence techniques with proposed hybrid feature selection.

Facial expressions are the most effective way of communication between humans. Hence, recognition of facial expressions is an emerging research topic in the area of image processing and pattern recognition. The task of recognizing facial expressions is challenging because of variation in many parameters like illumination, pose, ethnicity etc. This article presents an effective approach for recognition of facial expressions with variation in illumination. ROI based Local Binary Pattern is used for extracting feature information and Neural Network as classifier. Japanese Female Facial Expressions (JAFFE) database is used for obtaining the results.

In this modern era of technology control of leakage power consumption is a growing design challenge with scaling down in deep-submicron. This paper presents the DYNOC logic approach which has less leakage current with more stability with PVT parameters variation. Six PVT parameters of DYNOC circuit and conventional Domino logic circuit has been analyzed and compare using Tanner EDA tool at 70 nm technology node. Simulation results shows that the DYNOC logic approach has high voltage gain and gain margin with more temperature variation stability and also has improvement in leakage current reduction with variation in parameters like channel length, oxide thickness, threshold voltage size of PMOS to NMOS, supply voltage over Domino logic approach.

In this paper, an efficient fuzzy min-max neural network (EFMNN) method is proposed for classification to overcome the limitations of Simpson’s method (FMMNN). Classifier is basically union of fuzzy set hyperboxes. Algorithm comprises of three steps i.e. Expansion, Overlap and Contraction. The first input pattern coming to the classifier is not given same min and max values so that more points are accommodated in the created hyperbox and overlap criteria is recognized easily. In the expansion process next highest membership value is used when maximum membership hyperbox is bounded by maximum size specified by user to expand. Spiliting of bigger hyperbox is done to minimize the loss in area in contraction process. Experiments are done on standard datasets to compare the results of original FMMNN and EFMNN. The results empirically validated the usefulness of the modification proposed.

With the current population and growing number of vehicles on road, speed control has become an important requirement due to the increased number of accidents reported in our day-to-day life. Measures like traffic management, improving quality of road infrastructure and safer vehicles can prevent the road accidents. Though there are many technologies available for speed management of a vehicle, practical implementation of each is to some extent is not up to satisfactory level. This paper discusses a speed monitoring and control system that can provide safer environment in critical zones like schools, colleges etc. The proposed method uses Zigbee and GPS for the speed control and which is also cost effective and can be fixed easily on vehicles.

Many of the available researches have been concentrating on the profits maximization of cloud providers, while the actual necessities of cloud users have been ignored. Here, a Clustering based User preference (CUP) resource scheduling technique is proposed that can be used by a cloud provider for meeting the resource needs of a user in a better way. The novel CUP scheduling mechanism consists of four stages: resource matching, resource selection, clustering and resource scheduling. The user must be given a consideration if the same user puts forward multiple requirements. Updating of user demands and preferences are done at the resource scheduling stage. This method chooses the “best” VM which improves resourcefulness of CC and thereby minimizes the average response time of tasks. The results show that the CUP algorithm proposed efficiently satisfies the diverse requirements of the users and assists in the better resource utilization.

Gestures are very important for the communication between humans. Nowadays new technologies of Human Computer Interaction (HCI) are being developed to convey commands from user’s to computers. Among them, communication though hand gesture is a natural and instinctive way to interact with the computer. Over the years Vision based real time gesture recognition system has witnessed a large growth because of its different beneficial applications, such as sign language to virtual reality and its capability to interact with system efficiently through HCI. In HCI using hand gesture, the main elements are hand gesture recognition and its tracking. At first a brief history of hand gesture recognition is deliberated and the technical challenges during the recognition process also enumerated. The methodologies for hand gesture identification, such as glove-based, vision-based and depth-based, are compared briefly. Bare hand gesture recognition and tracking using CamShift can provide a better Human-Computer Interaction.

Data mining has its popularity among crime data analysis significantly due to increasing crime rates across the globe. In this research, classification methods are applied for predicting the nature of a crime that is whether the crime is a violent crime or a non-violent crime. In this work, we present two classification algorithms – Gradient Boosting algorithm and Random Forest algorithm for predicting the crime as a violent or non-violent crime and analyze the accuracy, precision and recall values of these algorithms for the crime records. The dataset is taken from the Communities and Crime data from UCI repository for processing. Further, to improve the accuracy of the predicted results, we use Boruta algorithm which is primarily a wrapper-algorithm for all relevant feature selections. The study finds that Boruta algorithm performs better in feature selection than the Chi-Square feature selection algorithm.

Now-a day’s remote users can get the benefit of different services from different servers using the internet. In such multi-server environment, one major security drawback is to identify the legitimate remote user of a web service on the internet. Traditional two factor authentication technique is vulnerable to, password guessing attack, stolen verifier attack, man-in-the-middle attack, etc. To eliminate this security problem, biometric authentication is essential. Keystroke dynamics, one of the behavioural biometric features can be used for remote user authentication. In this paper, a three factor based remote user authentication protocol has been proposed which uses keystroke dynamics as a third factor with two other previous factors. This authentication protocol will use machine learning techniques which results low false acceptance rate (FAR), false rejection rate (FRR) and low equal error rate (ERR).

For assisting blind or visually impaired persons, many computer vision technology has been developed. Some camera based systems were developed to help those people in way finding, navigation and finding daily necessities. The motion of the observer causes all scene object stationary or non-stationary in motion. And hence it is very much important to detect moving object with the moving observer. In this context we have proposed a camera based prototype system for assisting blind person in detection of obstacles by using motion vectors. We have collected dataset of their indoor and outdoor environment and estimated the optical flow to perform object detection. Furthermore we have detected the objects in the region of interest without using costly Depth cameras and sensors. The hardware used in the proposed work is ‘Raspberry Pi 2-B’ and the algorithms used for object detection is performed using MATLAB (for simulation purpose) and Python language.

Internet of things (IoT) has now become a fascinating system that improves information technology for its use in homes, cities and medical sectors. IoT works as an extension of internet to realize interconnections among every day object based on platform independent communication protocols. Object forming IoT must possess sensing, communication, and computation capabilities leading to a convenient as well as economical assistance for society. Interaction among heterogeneous objects enhances the security vulnerabilities in IoT. The current four layered communication stack of IoT supports protocols at each layer for enabling connectivity of heterogeneous objects. With security as a prime concern, communication in IoT need to maintained using a secure mechanism to protect the system from attacks. This paper analyzes various existing protocols at each layer with their inherent security mechanisms and exposing their vulnerabilities to different attacks. This paper will unlock new research areas for improving the inbuilt security mechanisms.

Sequential pattern mining is one of the most studied data mining problem and has wide range of application domains including weather prediction, network intrusion detection, web access analysis, customer purchase analysis, etc. The weighted sequential pattern mining is an approach to find only interesting sequential patterns by assigning weights to data elements present in the sequences. The time-interval weighted sequential pattern mining is another approach in which different weights are assigned to the time-interval values between the successive transactions. From customer purchase pattern analysis point of view, both item’s importance as well as time-interval gap values is useful and more interesting patterns can be discovered by considering them while assigning weights to the sequences. This paper aims to propose a novel approach for finding weighted sequential patterns from customer retail database which incorporates both the item’s importance and time-interval gap information so that the discovered sequential patterns will be more meaningful and effective for the end-user. The results infer a lot of computation cost can be saved by focusing on few interesting patterns.